LMCache is vulnerable to unauthenticated remote code execution via pickle deserialization in multiprocess mode transport. The vulnerability arises from an unauthenticated ZeroMQ ROUTER socket used for worker processes to register and share KV cache blocks in distributed mode. This socket lacks CURVE, ZAP, password, or message authentication. The transport binds to localhost by default unless configured to [truncated]
LMCache through 0.5.5 contains a server-side request forgery vulnerability in its frontend monitoring service. This allows unauthenticated attackers to bypass the proxy allowlist by registering arbitrary hosts, potentially leading to unauthorized access, data tampering, or disruption of node communication. The vulnerability can be exploited by adding entries via POST /api/proxies and then using /proxy or [truncated]
CVE-2026-107205: LMCache through 0.5.5 contains a missing authentication vulnerability in the multiprocess coordinator fleet control API. This allows remote unauthenticated attackers to access and manipulate caching and metadata, potentially disrupting operations and disclosing sensitive information. The vulnerability is highly severe with a CVSS score of 8.8. Defenders should assess exposure and potentia [truncated]
CVE-2026-107204 is a critical unauthenticated remote code execution vulnerability in LMCache through version 0.5.5. The vulnerability allows remote attackers to execute Python code by posting scripts to the /run_script endpoint. Attackers can recover real builtins through the injected FastAPI app object, bypassing the guarded __import__, to import os and run operating system commands as the LMCache proces [truncated]
A flaw has been found in LMCache up to 0.4.6. This affects the function hex_hash_to_int16 of the file lmcache/integration/vllm/utils.py of the component KV Cache Handler. Executing a manipulation can lead to use of weak hash. The attack needs to be launched locally. The attack requires a high level of complexity. It is indicated that the exploitability is difficult. The exploit has been published and may [truncated]
